includes/filerepo/backend/FileBackend.php

   1 <?php
   2 /**
   3  * @defgroup FileBackend File backend
   4  * @ingroup  FileRepo
   5  *
   6  * File backend is used to interact with file storage systems,
   7  * such as the local file system, NFS, or cloud storage systems.
   8  */
   9
  10 /**
  11  * Base class for all file backends.
  12  *
  13  * This program is free software; you can redistribute it and/or modify
  14  * it under the terms of the GNU General Public License as published by
  15  * the Free Software Foundation; either version 2 of the License, or
  16  * (at your option) any later version.
  17  *
  18  * This program is distributed in the hope that it will be useful,
  19  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  20  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
  21  * GNU General Public License for more details.
  22  *
  23  * You should have received a copy of the GNU General Public License along
  24  * with this program; if not, write to the Free Software Foundation, Inc.,
  25  * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
  26  * http://www.gnu.org/copyleft/gpl.html
  27  *
  28  * @file
  29  * @ingroup FileBackend
  30  * @author Aaron Schulz
  31  */
  32
  33 /**
  34  * @brief Base class for all file backend classes (including multi-write backends).
  35  *
  36  * This class defines the methods as abstract that subclasses must implement.
  37  * Outside callers can assume that all backends will have these functions.
  38  *
  39  * All "storage paths" are of the format "mwstore://<backend>/<container>/<path>".
  40  * The <path> portion is a relative path that uses UNIX file system (FS) notation,
  41  * though any particular backend may not actually be using a local filesystem.
  42  * Therefore, the relative paths are only virtual.
  43  *
  44  * Backend contents are stored under wiki-specific container names by default.
  45  * For legacy reasons, this has no effect for the FS backend class, and per-wiki
  46  * segregation must be done by setting the container paths appropriately.
  47  *
  48  * FS-based backends are somewhat more restrictive due to the existence of real
  49  * directory files; a regular file cannot have the same name as a directory. Other
  50  * backends with virtual directories may not have this limitation. Callers should
  51  * store files in such a way that no files and directories are under the same path.
  52  *
  53  * Methods should avoid throwing exceptions at all costs.
  54  * As a corollary, external dependencies should be kept to a minimum.
  55  *
  56  * @ingroup FileBackend
  57  * @since 1.19
  58  */
  59 abstract class FileBackend {
  60         protected $name; // string; unique backend name
  61         protected $wikiId; // string; unique wiki name
  62         protected $readOnly; // string; read-only explanation message
  63         protected $parallelize; // string; when to do operations in parallel
  64         protected $concurrency; // integer; how many operations can be done in parallel
  65
  66         /** @var LockManager */
  67         protected $lockManager;
  68         /** @var FileJournal */
  69         protected $fileJournal;
  70
  71         /**
  72          * Create a new backend instance from configuration.
  73          * This should only be called from within FileBackendGroup.
  74          *
  75          * $config includes:
  76          *     'name'        : The unique name of this backend.
  77          *                     This should consist of alphanumberic, '-', and '_' characters.
  78          *                     This name should not be changed after use.
  79          *     'wikiId'      : Prefix to container names that is unique to this wiki.
  80          *                     It should only consist of alphanumberic, '-', and '_' characters.
  81          *     'lockManager' : Registered name of a file lock manager to use.
  82          *     'fileJournal' : File journal configuration; see FileJournal::factory().
  83          *                     Journals simply log changes to files stored in the backend.
  84          *     'readOnly'    : Write operations are disallowed if this is a non-empty string.
  85          *                     It should be an explanation for the backend being read-only.
  86          *     'parallelize' : When to do file operations in parallel (when possible).
  87          *                     Allowed values are "implicit", "explicit" and "off".
  88          *     'concurrency' : How many file operations can be done in parallel.
  89          *
  90          * @param $config Array
  91          * @throws MWException
  92          */
  93         public function __construct( array $config ) {
  94                 $this->name = $config['name'];
  95                 if ( !preg_match( '!^[a-zA-Z0-9-_]{1,255}$!', $this->name ) ) {
  96                         throw new MWException( "Backend name `{$this->name}` is invalid." );
  97                 }
  98                 $this->wikiId = isset( $config['wikiId'] )
  99                         ? $config['wikiId']
 100                         : wfWikiID(); // e.g. "my_wiki-en_"
 101                 $this->lockManager = ( $config['lockManager'] instanceof LockManager )
 102                         ? $config['lockManager']
 103                         : LockManagerGroup::singleton()->get( $config['lockManager'] );
 104                 $this->fileJournal = isset( $config['fileJournal'] )
 105                         ? FileJournal::factory( $config['fileJournal'], $this->name )
 106                         : FileJournal::factory( array( 'class' => 'NullFileJournal' ), $this->name );
 107                 $this->readOnly = isset( $config['readOnly'] )
 108                         ? (string)$config['readOnly']
 109                         : '';
 110                 $this->parallelize = isset( $config['parallelize'] )
 111                         ? (string)$config['parallelize']
 112                         : 'off';
 113                 $this->concurrency = isset( $config['concurrency'] )
 114                         ? (int)$config['concurrency']
 115                         : 50;
 116         }
 117
 118         /**
 119          * Get the unique backend name.
 120          * We may have multiple different backends of the same type.
 121          * For example, we can have two Swift backends using different proxies.
 122          *
 123          * @return string
 124          */
 125         final public function getName() {
 126                 return $this->name;
 127         }
 128
 129         /**
 130          * Check if this backend is read-only
 131          *
 132          * @return bool
 133          */
 134         final public function isReadOnly() {
 135                 return ( $this->readOnly != '' );
 136         }
 137
 138         /**
 139          * Get an explanatory message if this backend is read-only
 140          *
 141          * @return string|bool Returns false if the backend is not read-only
 142          */
 143         final public function getReadOnlyReason() {
 144                 return ( $this->readOnly != '' ) ? $this->readOnly : false;
 145         }
 146
 147         /**
 148          * This is the main entry point into the backend for write operations.
 149          * Callers supply an ordered list of operations to perform as a transaction.
 150          * Files will be locked, the stat cache cleared, and then the operations attempted.
 151          * If any serious errors occur, all attempted operations will be rolled back.
 152          *
 153          * $ops is an array of arrays. The outer array holds a list of operations.
 154          * Each inner array is a set of key value pairs that specify an operation.
 155          *
 156          * Supported operations and their parameters:
 157          * a) Create a new file in storage with the contents of a string
 158          *     array(
 159          *         'op'                  => 'create',
 160          *         'dst'                 => <storage path>,
 161          *         'content'             => <string of new file contents>,
 162          *         'overwrite'           => <boolean>,
 163          *         'overwriteSame'       => <boolean>
 164          *     )
 165          * b) Copy a file system file into storage
 166          *     array(
 167          *         'op'                  => 'store',
 168          *         'src'                 => <file system path>,
 169          *         'dst'                 => <storage path>,
 170          *         'overwrite'           => <boolean>,
 171          *         'overwriteSame'       => <boolean>
 172          *     )
 173          * c) Copy a file within storage
 174          *     array(
 175          *         'op'                  => 'copy',
 176          *         'src'                 => <storage path>,
 177          *         'dst'                 => <storage path>,
 178          *         'overwrite'           => <boolean>,
 179          *         'overwriteSame'       => <boolean>
 180          *     )
 181          * d) Move a file within storage
 182          *     array(
 183          *         'op'                  => 'move',
 184          *         'src'                 => <storage path>,
 185          *         'dst'                 => <storage path>,
 186          *         'overwrite'           => <boolean>,
 187          *         'overwriteSame'       => <boolean>
 188          *     )
 189          * e) Delete a file within storage
 190          *     array(
 191          *         'op'                  => 'delete',
 192          *         'src'                 => <storage path>,
 193          *         'ignoreMissingSource' => <boolean>
 194          *     )
 195          * f) Do nothing (no-op)
 196          *     array(
 197          *         'op'                  => 'null',
 198          *     )
 199          *
 200          * Boolean flags for operations (operation-specific):
 201          * 'ignoreMissingSource' : The operation will simply succeed and do
 202          *                         nothing if the source file does not exist.
 203          * 'overwrite'           : Any destination file will be overwritten.
 204          * 'overwriteSame'       : An error will not be given if a file already
 205          *                         exists at the destination that has the same
 206          *                         contents as the new contents to be written there.
 207          *
 208          * $opts is an associative of boolean flags, including:
 209          * 'force'               : Operation precondition errors no longer trigger an abort.
 210          *                         Any remaining operations are still attempted. Unexpected
 211          *                         failures may still cause remaning operations to be aborted.
 212          * 'nonLocking'          : No locks are acquired for the operations.
 213          *                         This can increase performance for non-critical writes.
 214          *                         This has no effect unless the 'force' flag is set.
 215          * 'allowStale'          : Don't require the latest available data.
 216          *                         This can increase performance for non-critical writes.
 217          *                         This has no effect unless the 'force' flag is set.
 218          * 'nonJournaled'        : Don't log this operation batch in the file journal.
 219          *                         This limits the ability of recovery scripts.
 220          * 'parallelize'         : Try to do operations in parallel when possible.
 221          *
 222          * Remarks on locking:
 223          * File system paths given to operations should refer to files that are
 224          * already locked or otherwise safe from modification from other processes.
 225          * Normally these files will be new temp files, which should be adequate.
 226          *
 227          * Return value:
 228          * This returns a Status, which contains all warnings and fatals that occured
 229          * during the operation. The 'failCount', 'successCount', and 'success' members
 230          * will reflect each operation attempted. The status will be "OK" unless:
 231          *     a) unexpected operation errors occurred (network partitions, disk full...)
 232          *     b) significant operation errors occured and 'force' was not set
 233          *
 234          * @param $ops Array List of operations to execute in order
 235          * @param $opts Array Batch operation options
 236          * @return Status
 237          */
 238         final public function doOperations( array $ops, array $opts = array() ) {
 239                 if ( $this->isReadOnly() ) {
 240                         return Status::newFatal( 'backend-fail-readonly', $this->name, $this->readOnly );
 241                 }
 242                 if ( empty( $opts['force'] ) ) { // sanity
 243                         unset( $opts['nonLocking'] );
 244                         unset( $opts['allowStale'] );
 245                 }
 246                 $opts['concurrency'] = 1; // off
 247                 if ( $this->parallelize === 'implicit' ) {
 248                         if ( !isset( $opts['parallelize'] ) || $opts['parallelize'] ) {
 249                                 $opts['concurrency'] = $this->concurrency;
 250                         }
 251                 } elseif ( $this->parallelize === 'explicit' ) {
 252                         if ( !empty( $opts['parallelize'] ) ) {
 253                                 $opts['concurrency'] = $this->concurrency;
 254                         }
 255                 }
 256                 return $this->doOperationsInternal( $ops, $opts );
 257         }
 258
 259         /**
 260          * @see FileBackend::doOperations()
 261          */
 262         abstract protected function doOperationsInternal( array $ops, array $opts );
 263
 264         /**
 265          * Same as doOperations() except it takes a single operation.
 266          * If you are doing a batch of operations that should either
 267          * all succeed or all fail, then use that function instead.
 268          *
 269          * @see FileBackend::doOperations()
 270          *
 271          * @param $op Array Operation
 272          * @param $opts Array Operation options
 273          * @return Status
 274          */
 275         final public function doOperation( array $op, array $opts = array() ) {
 276                 return $this->doOperations( array( $op ), $opts );
 277         }
 278
 279         /**
 280          * Performs a single create operation.
 281          * This sets $params['op'] to 'create' and passes it to doOperation().
 282          *
 283          * @see FileBackend::doOperation()
 284          *
 285          * @param $params Array Operation parameters
 286          * @param $opts Array Operation options
 287          * @return Status
 288          */
 289         final public function create( array $params, array $opts = array() ) {
 290                 return $this->doOperation( array( 'op' => 'create' ) + $params, $opts );
 291         }
 292
 293         /**
 294          * Performs a single store operation.
 295          * This sets $params['op'] to 'store' and passes it to doOperation().
 296          *
 297          * @see FileBackend::doOperation()
 298          *
 299          * @param $params Array Operation parameters
 300          * @param $opts Array Operation options
 301          * @return Status
 302          */
 303         final public function store( array $params, array $opts = array() ) {
 304                 return $this->doOperation( array( 'op' => 'store' ) + $params, $opts );
 305         }
 306
 307         /**
 308          * Performs a single copy operation.
 309          * This sets $params['op'] to 'copy' and passes it to doOperation().
 310          *
 311          * @see FileBackend::doOperation()
 312          *
 313          * @param $params Array Operation parameters
 314          * @param $opts Array Operation options
 315          * @return Status
 316          */
 317         final public function copy( array $params, array $opts = array() ) {
 318                 return $this->doOperation( array( 'op' => 'copy' ) + $params, $opts );
 319         }
 320
 321         /**
 322          * Performs a single move operation.
 323          * This sets $params['op'] to 'move' and passes it to doOperation().
 324          *
 325          * @see FileBackend::doOperation()
 326          *
 327          * @param $params Array Operation parameters
 328          * @param $opts Array Operation options
 329          * @return Status
 330          */
 331         final public function move( array $params, array $opts = array() ) {
 332                 return $this->doOperation( array( 'op' => 'move' ) + $params, $opts );
 333         }
 334
 335         /**
 336          * Performs a single delete operation.
 337          * This sets $params['op'] to 'delete' and passes it to doOperation().
 338          *
 339          * @see FileBackend::doOperation()
 340          *
 341          * @param $params Array Operation parameters
 342          * @param $opts Array Operation options
 343          * @return Status
 344          */
 345         final public function delete( array $params, array $opts = array() ) {
 346                 return $this->doOperation( array( 'op' => 'delete' ) + $params, $opts );
 347         }
 348
 349         /**
 350          * Perform a set of independent file operations on some files.
 351          *
 352          * This does no locking, nor journaling, and possibly no stat calls.
 353          * Any destination files that already exist will be overwritten.
 354          * This should *only* be used on non-original files, like cache files.
 355          *
 356          * Supported operations and their parameters:
 357          * a) Create a new file in storage with the contents of a string
 358          *     array(
 359          *         'op'                  => 'create',
 360          *         'dst'                 => <storage path>,
 361          *         'content'             => <string of new file contents>
 362          *     )
 363          * b) Copy a file system file into storage
 364          *     array(
 365          *         'op'                  => 'store',
 366          *         'src'                 => <file system path>,
 367          *         'dst'                 => <storage path>
 368          *     )
 369          * c) Copy a file within storage
 370          *     array(
 371          *         'op'                  => 'copy',
 372          *         'src'                 => <storage path>,
 373          *         'dst'                 => <storage path>
 374          *     )
 375          * d) Move a file within storage
 376          *     array(
 377          *         'op'                  => 'move',
 378          *         'src'                 => <storage path>,
 379          *         'dst'                 => <storage path>
 380          *     )
 381          * e) Delete a file within storage
 382          *     array(
 383          *         'op'                  => 'delete',
 384          *         'src'                 => <storage path>,
 385          *         'ignoreMissingSource' => <boolean>
 386          *     )
 387          * f) Do nothing (no-op)
 388          *     array(
 389          *         'op'                  => 'null',
 390          *     )
 391          *
 392          * Boolean flags for operations (operation-specific):
 393          * 'ignoreMissingSource' : The operation will simply succeed and do
 394          *                         nothing if the source file does not exist.
 395          *
 396          * Return value:
 397          * This returns a Status, which contains all warnings and fatals that occured
 398          * during the operation. The 'failCount', 'successCount', and 'success' members
 399          * will reflect each operation attempted for the given files. The status will be
 400          * considered "OK" as long as no fatal errors occured.
 401          *
 402          * @param $ops Array Set of operations to execute
 403          * @return Status
 404          */
 405         final public function doQuickOperations( array $ops ) {
 406                 if ( $this->isReadOnly() ) {
 407                         return Status::newFatal( 'backend-fail-readonly', $this->name, $this->readOnly );
 408                 }
 409                 foreach ( $ops as &$op ) {
 410                         $op['overwrite'] = true; // avoids RTTs in key/value stores
 411                 }
 412                 return $this->doQuickOperationsInternal( $ops );
 413         }
 414
 415         /**
 416          * @see FileBackend::doQuickOperations()
 417          */
 418         abstract protected function doQuickOperationsInternal( array $ops );
 419
 420         /**
 421          * Concatenate a list of storage files into a single file system file.
 422          * The target path should refer to a file that is already locked or
 423          * otherwise safe from modification from other processes. Normally,
 424          * the file will be a new temp file, which should be adequate.
 425          * $params include:
 426          *     srcs : ordered source storage paths (e.g. chunk1, chunk2, ...)
 427          *     dst  : file system path to 0-byte temp file
 428          *
 429          * @param $params Array Operation parameters
 430          * @return Status
 431          */
 432         abstract public function concatenate( array $params );
 433
 434         /**
 435          * Prepare a storage directory for usage.
 436          * This will create any required containers and parent directories.
 437          * Backends using key/value stores only need to create the container.
 438          *
 439          * $params include:
 440          *     dir : storage directory
 441          *
 442          * @param $params Array
 443          * @return Status
 444          */
 445         final public function prepare( array $params ) {
 446                 if ( $this->isReadOnly() ) {
 447                         return Status::newFatal( 'backend-fail-readonly', $this->name, $this->readOnly );
 448                 }
 449                 return $this->doPrepare( $params );
 450         }
 451
 452         /**
 453          * @see FileBackend::prepare()
 454          */
 455         abstract protected function doPrepare( array $params );
 456
 457         /**
 458          * Take measures to block web access to a storage directory and
 459          * the container it belongs to. FS backends might add .htaccess
 460          * files whereas key/value store backends might restrict container
 461          * access to the auth user that represents end-users in web request.
 462          * This is not guaranteed to actually do anything.
 463          *
 464          * $params include:
 465          *     dir       : storage directory
 466          *     noAccess  : try to deny file access
 467          *     noListing : try to deny file listing
 468          *
 469          * @param $params Array
 470          * @return Status
 471          */
 472         final public function secure( array $params ) {
 473                 if ( $this->isReadOnly() ) {
 474                         return Status::newFatal( 'backend-fail-readonly', $this->name, $this->readOnly );
 475                 }
 476                 $status = $this->doPrepare( $params ); // dir must exist to restrict it
 477                 if ( $status->isOK() ) {
 478                         $status->merge( $this->doSecure( $params ) );
 479                 }
 480                 return $status;
 481         }
 482
 483         /**
 484          * @see FileBackend::secure()
 485          */
 486         abstract protected function doSecure( array $params );
 487
 488         /**
 489          * Delete a storage directory if it is empty.
 490          * Backends using key/value stores may do nothing unless the directory
 491          * is that of an empty container, in which case it should be deleted.
 492          *
 493          * $params include:
 494          *     dir       : storage directory
 495          *     recursive : recursively delete empty subdirectories first (@since 1.20)
 496          *
 497          * @param $params Array
 498          * @return Status
 499          */
 500         final public function clean( array $params ) {
 501                 if ( $this->isReadOnly() ) {
 502                         return Status::newFatal( 'backend-fail-readonly', $this->name, $this->readOnly );
 503                 }
 504                 return $this->doClean( $params );
 505         }
 506
 507         /**
 508          * @see FileBackend::clean()
 509          */
 510         abstract protected function doClean( array $params );
 511
 512         /**
 513          * Check if a file exists at a storage path in the backend.
 514          * This returns false if only a directory exists at the path.
 515          *
 516          * $params include:
 517          *     src    : source storage path
 518          *     latest : use the latest available data
 519          *
 520          * @param $params Array
 521          * @return bool|null Returns null on failure
 522          */
 523         abstract public function fileExists( array $params );
 524
 525         /**
 526          * Get the last-modified timestamp of the file at a storage path.
 527          *
 528          * $params include:
 529          *     src    : source storage path
 530          *     latest : use the latest available data
 531          *
 532          * @param $params Array
 533          * @return string|bool TS_MW timestamp or false on failure
 534          */
 535         abstract public function getFileTimestamp( array $params );
 536
 537         /**
 538          * Get the contents of a file at a storage path in the backend.
 539          * This should be avoided for potentially large files.
 540          *
 541          * $params include:
 542          *     src    : source storage path
 543          *     latest : use the latest available data
 544          *
 545          * @param $params Array
 546          * @return string|bool Returns false on failure
 547          */
 548         abstract public function getFileContents( array $params );
 549
 550         /**
 551          * Get the size (bytes) of a file at a storage path in the backend.
 552          *
 553          * $params include:
 554          *     src    : source storage path
 555          *     latest : use the latest available data
 556          *
 557          * @param $params Array
 558          * @return integer|bool Returns false on failure
 559          */
 560         abstract public function getFileSize( array $params );
 561
 562         /**
 563          * Get quick information about a file at a storage path in the backend.
 564          * If the file does not exist, then this returns false.
 565          * Otherwise, the result is an associative array that includes:
 566          *     mtime  : the last-modified timestamp (TS_MW)
 567          *     size   : the file size (bytes)
 568          * Additional values may be included for internal use only.
 569          *
 570          * $params include:
 571          *     src    : source storage path
 572          *     latest : use the latest available data
 573          *
 574          * @param $params Array
 575          * @return Array|bool|null Returns null on failure
 576          */
 577         abstract public function getFileStat( array $params );
 578
 579         /**
 580          * Get a SHA-1 hash of the file at a storage path in the backend.
 581          *
 582          * $params include:
 583          *     src    : source storage path
 584          *     latest : use the latest available data
 585          *
 586          * @param $params Array
 587          * @return string|bool Hash string or false on failure
 588          */
 589         abstract public function getFileSha1Base36( array $params );
 590
 591         /**
 592          * Get the properties of the file at a storage path in the backend.
 593          * Returns FSFile::placeholderProps() on failure.
 594          *
 595          * $params include:
 596          *     src    : source storage path
 597          *     latest : use the latest available data
 598          *
 599          * @param $params Array
 600          * @return Array
 601          */
 602         abstract public function getFileProps( array $params );
 603
 604         /**
 605          * Stream the file at a storage path in the backend.
 606          * If the file does not exists, a 404 error will be given.
 607          * Appropriate HTTP headers (Status, Content-Type, Content-Length)
 608          * must be sent if streaming began, while none should be sent otherwise.
 609          * Implementations should flush the output buffer before sending data.
 610          *
 611          * $params include:
 612          *     src     : source storage path
 613          *     headers : additional HTTP headers to send on success
 614          *     latest  : use the latest available data
 615          *
 616          * @param $params Array
 617          * @return Status
 618          */
 619         abstract public function streamFile( array $params );
 620
 621         /**
 622          * Returns a file system file, identical to the file at a storage path.
 623          * The file returned is either:
 624          * a) A local copy of the file at a storage path in the backend.
 625          *    The temporary copy will have the same extension as the source.
 626          * b) An original of the file at a storage path in the backend.
 627          * Temporary files may be purged when the file object falls out of scope.
 628          *
 629          * Write operations should *never* be done on this file as some backends
 630          * may do internal tracking or may be instances of FileBackendMultiWrite.
 631          * In that later case, there are copies of the file that must stay in sync.
 632          * Additionally, further calls to this function may return the same file.
 633          *
 634          * $params include:
 635          *     src    : source storage path
 636          *     latest : use the latest available data
 637          *
 638          * @param $params Array
 639          * @return FSFile|null Returns null on failure
 640          */
 641         abstract public function getLocalReference( array $params );
 642
 643         /**
 644          * Get a local copy on disk of the file at a storage path in the backend.
 645          * The temporary copy will have the same file extension as the source.
 646          * Temporary files may be purged when the file object falls out of scope.
 647          *
 648          * $params include:
 649          *     src    : source storage path
 650          *     latest : use the latest available data
 651          *
 652          * @param $params Array
 653          * @return TempFSFile|null Returns null on failure
 654          */
 655         abstract public function getLocalCopy( array $params );
 656
 657         /**
 658          * Check if a directory exists at a given storage path.
 659          * Backends using key/value stores will check if the path is a
 660          * virtual directory, meaning there are files under the given directory.
 661          *
 662          * Storage backends with eventual consistency might return stale data.
 663          *
 664          * $params include:
 665          *     dir : storage directory
 666          *
 667          * @param $params array
 668          * @return bool|null Returns null on failure
 669          * @since 1.20
 670          */
 671         abstract public function directoryExists( array $params );
 672
 673         /**
 674          * Get an iterator to list *all* directories under a storage directory.
 675          * If the directory is of the form "mwstore://backend/container",
 676          * then all directories in the container should be listed.
 677          * If the directory is of form "mwstore://backend/container/dir",
 678          * then all directories directly under that directory should be listed.
 679          * Results should be storage directories relative to the given directory.
 680          *
 681          * Storage backends with eventual consistency might return stale data.
 682          *
 683          * $params include:
 684          *     dir     : storage directory
 685          *     topOnly : only return direct child dirs of the directory
 686          *
 687          * @param $params array
 688          * @return Traversable|Array|null Returns null on failure
 689          * @since 1.20
 690          */
 691         abstract public function getDirectoryList( array $params );
 692
 693         /**
 694          * Same as FileBackend::getDirectoryList() except only lists
 695          * directories that are immediately under the given directory.
 696          *
 697          * Storage backends with eventual consistency might return stale data.
 698          *
 699          * $params include:
 700          *     dir : storage directory
 701          *
 702          * @param $params array
 703          * @return Traversable|Array|null Returns null on failure
 704          * @since 1.20
 705          */
 706         final public function getTopDirectoryList( array $params ) {
 707                 return $this->getDirectoryList( array( 'topOnly' => true ) + $params );
 708         }
 709
 710         /**
 711          * Get an iterator to list *all* stored files under a storage directory.
 712          * If the directory is of the form "mwstore://backend/container",
 713          * then all files in the container should be listed.
 714          * If the directory is of form "mwstore://backend/container/dir",
 715          * then all files under that directory should be listed.
 716          * Results should be storage paths relative to the given directory.
 717          *
 718          * Storage backends with eventual consistency might return stale data.
 719          *
 720          * $params include:
 721          *     dir     : storage directory
 722          *     topOnly : only return direct child files of the directory (@since 1.20)
 723          *
 724          * @param $params array
 725          * @return Traversable|Array|null Returns null on failure
 726          */
 727         abstract public function getFileList( array $params );
 728
 729         /**
 730          * Same as FileBackend::getFileList() except only lists
 731          * files that are immediately under the given directory.
 732          *
 733          * Storage backends with eventual consistency might return stale data.
 734          *
 735          * $params include:
 736          *     dir : storage directory
 737          *
 738          * @param $params array
 739          * @return Traversable|Array|null Returns null on failure
 740          * @since 1.20
 741          */
 742         final public function getTopFileList( array $params ) {
 743                 return $this->getFileList( array( 'topOnly' => true ) + $params );
 744         }
 745
 746         /**
 747          * Invalidate any in-process file existence and property cache.
 748          * If $paths is given, then only the cache for those files will be cleared.
 749          *
 750          * @param $paths Array Storage paths (optional)
 751          * @return void
 752          */
 753         public function clearCache( array $paths = null ) {}
 754
 755         /**
 756          * Lock the files at the given storage paths in the backend.
 757          * This will either lock all the files or none (on failure).
 758          *
 759          * Callers should consider using getScopedFileLocks() instead.
 760          *
 761          * @param $paths Array Storage paths
 762          * @param $type integer LockManager::LOCK_* constant
 763          * @return Status
 764          */
 765         final public function lockFiles( array $paths, $type ) {
 766                 return $this->lockManager->lock( $paths, $type );
 767         }
 768
 769         /**
 770          * Unlock the files at the given storage paths in the backend.
 771          *
 772          * @param $paths Array Storage paths
 773          * @param $type integer LockManager::LOCK_* constant
 774          * @return Status
 775          */
 776         final public function unlockFiles( array $paths, $type ) {
 777                 return $this->lockManager->unlock( $paths, $type );
 778         }
 779
 780         /**
 781          * Lock the files at the given storage paths in the backend.
 782          * This will either lock all the files or none (on failure).
 783          * On failure, the status object will be updated with errors.
 784          *
 785          * Once the return value goes out scope, the locks will be released and
 786          * the status updated. Unlock fatals will not change the status "OK" value.
 787          *
 788          * @param $paths Array Storage paths
 789          * @param $type integer LockManager::LOCK_* constant
 790          * @param $status Status Status to update on lock/unlock
 791          * @return ScopedLock|null Returns null on failure
 792          */
 793         final public function getScopedFileLocks( array $paths, $type, Status $status ) {
 794                 return ScopedLock::factory( $this->lockManager, $paths, $type, $status );
 795         }
 796
 797         /**
 798          * Get an array of scoped locks needed for a batch of file operations.
 799          *
 800          * Normally, FileBackend::doOperations() handles locking, unless
 801          * the 'nonLocking' param is passed in. This function is useful if you
 802          * want the files to be locked for a broader scope than just when the
 803          * files are changing. For example, if you need to update DB metadata,
 804          * you may want to keep the files locked until finished.
 805          *
 806          * @see FileBackend::doOperations()
 807          *
 808          * @param $ops Array List of file operations to FileBackend::doOperations()
 809          * @param $status Status Status to update on lock/unlock
 810          * @return Array List of ScopedFileLocks or null values
 811          * @since 1.20
 812          */
 813         abstract public function getScopedLocksForOps( array $ops, Status $status );
 814
 815         /**
 816          * Get the root storage path of this backend.
 817          * All container paths are "subdirectories" of this path.
 818          *
 819          * @return string Storage path
 820          * @since 1.20
 821          */
 822         final public function getRootStoragePath() {
 823                 return "mwstore://{$this->name}";
 824         }
 825
 826         /**
 827          * Get the file journal object for this backend
 828          *
 829          * @return FileJournal
 830          */
 831         final public function getJournal() {
 832                 return $this->fileJournal;
 833         }
 834
 835         /**
 836          * Check if a given path is a "mwstore://" path.
 837          * This does not do any further validation or any existence checks.
 838          *
 839          * @param $path string
 840          * @return bool
 841          */
 842         final public static function isStoragePath( $path ) {
 843                 return ( strpos( $path, 'mwstore://' ) === 0 );
 844         }
 845
 846         /**
 847          * Split a storage path into a backend name, a container name,
 848          * and a relative file path. The relative path may be the empty string.
 849          * This does not do any path normalization or traversal checks.
 850          *
 851          * @param $storagePath string
 852          * @return Array (backend, container, rel object) or (null, null, null)
 853          */
 854         final public static function splitStoragePath( $storagePath ) {
 855                 if ( self::isStoragePath( $storagePath ) ) {
 856                         // Remove the "mwstore://" prefix and split the path
 857                         $parts = explode( '/', substr( $storagePath, 10 ), 3 );
 858                         if ( count( $parts ) >= 2 && $parts[0] != '' && $parts[1] != '' ) {
 859                                 if ( count( $parts ) == 3 ) {
 860                                         return $parts; // e.g. "backend/container/path"
 861                                 } else {
 862                                         return array( $parts[0], $parts[1], '' ); // e.g. "backend/container"
 863                                 }
 864                         }
 865                 }
 866                 return array( null, null, null );
 867         }
 868
 869         /**
 870          * Normalize a storage path by cleaning up directory separators.
 871          * Returns null if the path is not of the format of a valid storage path.
 872          *
 873          * @param $storagePath string
 874          * @return string|null
 875          */
 876         final public static function normalizeStoragePath( $storagePath ) {
 877                 list( $backend, $container, $relPath ) = self::splitStoragePath( $storagePath );
 878                 if ( $relPath !== null ) { // must be for this backend
 879                         $relPath = self::normalizeContainerPath( $relPath );
 880                         if ( $relPath !== null ) {
 881                                 return ( $relPath != '' )
 882                                         ? "mwstore://{$backend}/{$container}/{$relPath}"
 883                                         : "mwstore://{$backend}/{$container}";
 884                         }
 885                 }
 886                 return null;
 887         }
 888
 889         /**
 890          * Get the parent storage directory of a storage path.
 891          * This returns a path like "mwstore://backend/container",
 892          * "mwstore://backend/container/...", or null if there is no parent.
 893          *
 894          * @param $storagePath string
 895          * @return string|null
 896          */
 897         final public static function parentStoragePath( $storagePath ) {
 898                 $storagePath = dirname( $storagePath );
 899                 list( $b, $cont, $rel ) = self::splitStoragePath( $storagePath );
 900                 return ( $rel === null ) ? null : $storagePath;
 901         }
 902
 903         /**
 904          * Get the final extension from a storage or FS path
 905          *
 906          * @param $path string
 907          * @return string
 908          */
 909         final public static function extensionFromPath( $path ) {
 910                 $i = strrpos( $path, '.' );
 911                 return strtolower( $i ? substr( $path, $i + 1 ) : '' );
 912         }
 913
 914         /**
 915          * Check if a relative path has no directory traversals
 916          *
 917          * @param $path string
 918          * @return bool
 919          * @since 1.20
 920          */
 921         final public static function isPathTraversalFree( $path ) {
 922                 return ( self::normalizeContainerPath( $path ) !== null );
 923         }
 924
 925         /**
 926          * Validate and normalize a relative storage path.
 927          * Null is returned if the path involves directory traversal.
 928          * Traversal is insecure for FS backends and broken for others.
 929          *
 930          * This uses the same traversal protection as Title::secureAndSplit().
 931          *
 932          * @param $path string Storage path relative to a container
 933          * @return string|null
 934          */
 935         final protected static function normalizeContainerPath( $path ) {
 936                 // Normalize directory separators
 937                 $path = strtr( $path, '\\', '/' );
 938                 // Collapse any consecutive directory separators
 939                 $path = preg_replace( '![/]{2,}!', '/', $path );
 940                 // Remove any leading directory separator
 941                 $path = ltrim( $path, '/' );
 942                 // Use the same traversal protection as Title::secureAndSplit()
 943                 if ( strpos( $path, '.' ) !== false ) {
 944                         if (
 945                                 $path === '.' ||
 946                                 $path === '..' ||
 947                                 strpos( $path, './' ) === 0 ||
 948                                 strpos( $path, '../' ) === 0 ||
 949                                 strpos( $path, '/./' ) !== false ||
 950                                 strpos( $path, '/../' ) !== false
 951                         ) {
 952                                 return null;
 953                         }
 954                 }
 955                 return $path;
 956         }
 957 }